Weak Edge Detection Research Articles

Edge Detection of COVID-19 CT Image Based on GF_SSR, Improved Multiscale Morphology, and Adaptive Threshold

Edge detection is an effective method for image segmentation and feature extraction. Therefore, extracting weak edges with the inhomogeneous gray of Corona Virus Disease 2019 (COVID-19) CT images is extremely important. Multiscale morphology has been widely used in the edge detection of medical images due to its excellent boundary detection accuracy. In this paper, we propose a weak edge detection method based on Gaussian filtering and singlescale Retinex (GF_SSR), and improved multiscale morphology and adaptive threshold binarization (IMSM_ATB). As all the CT images have noise, we propose to remove image noise by Gaussian filtering. The edge of CT images is enhanced using the SSR algorithm. In addition, based on the extracted edge of CT images using improved Multiscale morphology, a particle swarm optimization (PSO) algorithm is introduced to binarize the image by automatically getting the optimal threshold. To evaluate our method, we use images from three datasets, namely COVID-19, Kaggle-COVID-19, and COVID-Chestxray, respectively. The average values of results are worthy of reference, with the Shannon information entropy of 1.8539, the Precision of 0.9992, the Recall of 0.8224, the F-Score of 1.9158, running time of 11.3000. Finally, three types of lesion images in the COVID-19 dataset are selected to evaluate the visual effects of the proposed algorithm. Compared with the other four algorithms, the proposed algorithm effectively detects the weak edge of the lesion and provides help for image segmentation and feature extraction.

A Study on Weak Edge Detection of COVID-19's CT Images Based on Histogram Equalization and Improved Canny Algorithm.

The coronavirus disease 2019 (COVID-19) is a substantial threat to people's lives and health due to its high infectivity and rapid spread. Computed tomography (CT) scan is one of the important auxiliary methods for the clinical diagnosis of COVID-19. However, CT image lesion edge is normally affected by pixels with uneven grayscale and isolated noise, which makes weak edge detection of the COVID-19 lesion more complicated. In order to solve this problem, an edge detection method is proposed, which combines the histogram equalization and the improved Canny algorithm. Specifically, the histogram equalization is applied to enhance image contrast. In the improved Canny algorithm, the median filter, instead of the Gaussian filter, is used to remove the isolated noise points. The K-means algorithm is applied to separate the image background and edge. And the Canny algorithm is improved continuously by combining the mathematical morphology and the maximum between class variance method (OTSU). On selecting four types of lesion images from COVID-CT date set, MSE, MAE, SNR, and the running time are applied to evaluate the performance of the proposed method. The average values of these evaluation indicators are 1.7322, 7.9010, 57.1241, and 5.4887, respectively. Compared with other three methods, these values indicate that the proposed method achieves better result. The experimental results prove that the proposed algorithm can effectively detect the weak edge of the lesion, which is helpful for the diagnosis of COVID-19.

Improved ROEWA SAR Image Edge Detector Based on Curvilinear Structures Extraction

By introducing curvilinear structures extraction (CSE) instead of watershed algorithm (WA) or nonmaximum suppression (NMS) to edge strength map (ESM), an improved ratio of exponentially weighted averages (ROEWA) edge detector with better capacity for weak edges detection is proposed to extract smooth edges and edge direction of synthetic aperture radar (SAR) images. Using the ROEWA, the ESM is calculated. Then the CSE algorithm is employed to extracted edges, by acquiring eigenvectors and eigenvalues of the Hessian matrix, the improved ESM (IESM) is obtained, which ensures the good weak edge detection capacity and smoothness of edges. Experimental results on simulated and real SAR images show that the improved ROEWA based on CSE attains better performance than the one using WA or NMS.

Segmentation of high-resolution remote sensing image combining phase congruency with local homogeneity

Realizing both of the effective weak edge detection and fake edge suppression is an extremely challenging problem facing high-resolution remote sensing (HRRS) image segmentation. To address the problem, an HRRS image segmentation method combining phase congruency with local homogeneity is proposed by advantageous complementarities. Firstly, the Log Gabor Filter is used to extract phase congruency information. Then, the local homogeneity index J value is adopted to optimize the edge detection results. On this basis, an objective function optimization strategy based on minimizing inter-scale mutual information is proposed, and a parameter-adaptive model of edge response is established. In the end, the segmentation results are obtained by multi-scale region segmentation and merging based on this model. Two sets of HRRS images are used in experiments, and the results are compared with the J value-/phase congruency-based models, the well-known commercial software e-Cognition, and a traditional gradient-based segmentation method, respectively. Both visual and quantitative evaluations have demonstrated the effectiveness of the proposed method.

A Comparative Study on Left and Right Endocardium Segmentation using Gradient Vector Field and Adaptive Diffusion Flow Algorithms

The cardiac Magnetic Resonance Imaging (MRI) provides high resolution images of the heart without radiation exposure. It is an excellent noninvasive test used by radiologist for proper detection of heart diseases. The manual segmentation of left ventricle in cine short axis MRI sequences takes an ample amount of time as compared to semi-automated segmentation. In Gradient Vector Flow (GVF) model certain barriers hinder the performance such as weak edge detection, high computational time, limited capture range and its ambiguity with other parameters. In this paper segmentation of Endocardium is carried out on multistage MRI frames Using Adaptive Diffusion flow (ADF) model. This deformable model was tested on large scale number of Cardiac MRI images. We replace the smoothening energy term in GVF with active hyper-surface harmonic minimal function in order to avoid possible leakage at weak edges. The use of harmonic maps is adjusted in accordance with image characteristics. We also assimilate infinite Laplace function to move active contours into narrow concave sections. Experimental results and collation with GVF are presented in this paper which reveals several good results based on extraction of endocardium tissue from left and right ventricle, including less computational time, noise robustness and weak edge preserving on Cardiac MR Images.

Weak edge detection of CT image based on improved algorithm of fuzzy theory

In order to solve the weak edge detection of traditional industrial CT images shortcomings of poor detection effect and low speed, a detection method of step fuzzy inference algorithm and another method of improved defuzzification algorithm were researched. Compared with overall reasoning method, the step fuzzy inference algorithm selected similarity, gradient and consistency as blur characteristics. And the reasoning process used Mandani reasoning to conduct step fuzzy reasoning, which was based on simplified inference rule tables. Improved defuzzification algorithm was proposed in the solution process. According to membership function figure, the two methods were verified by experiments. The results showed that the step fuzzy inference algorithm was better in the weak edge detection, while the improved defuzzification algorithm greatly increased the computing speed on the premise of accuracy.

模糊理论改进算法的CT图像弱边缘检测

模糊理论改进算法的CT图像弱边缘检测

Investigations of Cellular Automata Linear Rules for Edge Detection

Edge detection of images is one of the basic and most significant operations in image processing and is used for object background separation, 3-D interpretation of a 2-D image, and pre-processing in image understanding and recognition algorithms. In this paper we investigate cellular automata linear rules for edge detection and based on this investigation we have classified the rules into no edge detection rules, strong edge detection rules and weak edge detection rules. Finally, we show the comparative analysis of the proposed technique with already defined techniques for edge detection and the results show desirable performance.

Algorithm for the Detection of Feeble Edge of Complex Image

Image weak edge detection has a wide application. An improved ant colony algorithm is used for image edge detection . When image background is very complex ,the weak edge detection is more difficult. Around the key steps in the existing ant colony algorithm , we propose two improved methods. First, the expression of pheromone has been improved, and second, to improve the heuristic information value is calculated. Finally, the improved ant colony algorithm with the traditional Canny edge detection algorithm are compared.The improved edge detection experiments show that the new algorithm is position detection accurate, continuity, and the advantages of less interference.

A Study of Image Weak-Edge Detection Combination of BP Neural Networks

This paper proposes an image weak-edge detection method based on the combination of edge features and BP neural networks. Through analyzing the basic characteristics of the image edge points, we construct 8 groups 3-D feature vectors as the training sample set, combining with the learning function based on gradient descent momentum and the Levenberg-Marquardt training function, to train the BP neural network, further complete the image edge detection. Finally, compared with the traditional edge detection methods, the experimental results show that this method can detect the weak-edge and corner-edge much better.

The construction of labeled line drawings from intensity images

This paper describes the POLL (Perceptual Organization and Line Labeling) system for obtaining labeled line drawings from single intensity images using an integrated blackboard system. This system emphasizes the data driven extraction of 3D geometric information from intensity images. The robustness of the system comes from it being implemented in an integrated framework in which the errors made by one module can be diagnosed and corrected by the constraints imposed by other modules. The system is able to generate an initial line drawing from an intensity image in the domain of piecewise objects. Four modules were used as knowledge sources: weak membrance edge detection, curvilinear grouping, proximity grouping and curvilinear line labeling. An initial representation of the image data is built using the first three knowledge sources. This representation is analyzed using a modified curvilinear line labeling algorithm developed in this paper that uses figure-ground separation to constrain legal line labelings. This modified line labeling algorithm can diagnose problems with the initial representation. The modified line labeling algorithm can find errors such as missing edges, improperly typed verticles, and missing phantom junctions. Errors in the representation can be fixed using a set of heuristics that were created to repair common mistakes. If no irreparable errors are found in the representation, then the modified line labeling algorithm produces a 3D interpretation of the data in the input image without explicitly reconstructing 3D shape.